Search

Dispersal kernels are the standard method in biology for describing and predicting the relationship between dispersal and distance. Statistically fitted dispersal kernels allow observations of a limited number of dispersal events to be extrapolated across a wider landscape, and form the basis of a wide range of theories and methods in ecology, evolution and conservation. Genetic parentage data are an increasingly common source of dispersal information, particularly for species where dispersal is difficult to observe directly. In particular, parentage analysis is now routinely applied to coral reef fish, whose larvae can potentially disperse over many kilometres, and are too small to track in situ. It is not straightforward to estimate dispersal kernels from parentage data, and existing methods all have substantial limitations. These include the omission of important population processes such as density-dependent mortality, and data on unassigned juveniles. Here we develop and proof a new likelihood estimator for fitting dispersal kernels to parentage data, applying it to simulated parentage datasets for coral reef fish on the Great Barrier Reef (GBR). The method incorporates a series of factors not previously considered in other methods: the partial sampling of adults and juveniles on sampled sites; the existence of unassigned dispersers from unsampled habitat patches; and post-settlement processes (e.g. density-dependent mortality) that follow dispersal but precede parentage sampling. Including these additional factors requires an estimate of adult populations on unsampled habitat patches, but the result is a superior estimate of dispersal kernels and mean dispersal distances. Our power analyses suggest that the parentage datasets currently available for reef fishes are large enough to fit accurate dispersal kernels. Based on the analyses in one particular region of the GBR, parentage sampling should be distributed equally between adults and juveniles, and should s

To contribute to the aspirations of recent international biodiversity conventions, protected areas (PAs) must be strategically located and not simply established on economically marginal lands as they have in the past. With refined international commitments under the Convention on Biological Diversity to target protected areas in places of “importance to biodiversity,” perhaps they may now be. We analyzed location biases in PAs globally over historic (pre-2004) and recent periods. Specifically, we examined whether the location of protected areas are more closely associated with high concentrations of threatened vertebrate species or with areas of low agricultural opportunity costs. We found that both old and new protected areas did not target places with high concentrations of threatened vertebrate species. Instead, they appeared to be established in locations that minimize conflict with agriculturally suitable lands. This entrenchment of past trends has substantial implications for the contributions these protected areas are making to international commitments to conserve biodiversity. If protected-area growth from 2004 to 2014 had strategically targeted unrepresented threatened vertebrates, >30 times more species (3086 or 2553 potential vs. 85 actual new species represented) would have been protected for the same area or the same cost as the actual expansion. With the land available for conservation declining, nations must urgently focus new protection on places that provide for the conservation outcomes outlined in international treaties.